Identification of the PA1113 Gene Product as an ABC Transporter Involved in the Uptake of Carbenicillin in Pseudomonas aeruginosa PAO1

Abstract

The resistance of Pseudomonas aeruginosa to antibiotics is multi factorial and complex. Whereas efflux pumps such as MexAB-OprM have been thought to predominate, here we show that a novel ATP Binding Cassette (ABC) transporter that mediates influx of carbenicillin from the periplasm to the cytoplasm and away from its cell wall target plays an important role in the resistance of P. aeruginosa to this antibiotic. Treatment of P. aeruginosa with verapamil, an inhibitor of ABC transporters in eukaryotic cells, increases its sensitivity to carbenicillin. Using amino acid sequence homology with known verapamil protein targets as a probe, we determined that the PA1113 gene product, an ABC transporter, mediates carbenicillin uptake into the bacterial cytoplasm. Docking and pharmacological analyses showed that verapamil and carbenicillin compete for the same site on the PA1113 gene protein, explaining the inhibitory effect of verapamil on carbenicillin uptake, and furthermore suggest that the PA1113 ABC transporter accounts for about 30% of P. aeruginosa carbenicillin resistance. Our findings demonstrate that the PA1113 gene product helps mediate carbenicillin resistance by transporting it away from its cell wall target and represents a promising new therapeutic target.

Keywords: ABC transporter; Pseudomonas. aeruginosa; carbenicillin; verapamil.

Figure 1

Effects of verapamil on P. aeruginosa growth and its sensitivity to carbenicillin.

Figure 2

Effects of verapamil on sensitivity of several PAO1 mutants to carbenicillin. P. aeruginosa PAO1 mutants Pseudomonas Washington (PW) mutant PW3010, PW6408, and PW8020 were inoculated in 96 well plates with increasing concentrations in verapamil and carbenicillin. Growths and IC₅₀ values were measured as described in the legend of Figure 1. The IC₅₀ values were plotted against concentration in verapamil: PW8020 ■, PW6408 ▲, PW3010 ●. Results presented were the mean values of at least four independent experiments.

Figure 3

Expression of the PA1113 gene in PAO1 and its mutant PW3010. (Top) Schematic representation of the PA1113 gene. +136 bp after the ATG, shows the insertion position of the transposon (ISlacZ/hah). Below are the amplicons that were synthesized by RT-PCR to check the expression of the gene. (Bottom) The results generated by PCR from cDNAs obtained after reverse transcription on the total RNAs of PAO1 (on the left) and its mutant (on the right). RT-PCR was performed on total RNAs as described in the Materials and Methods section.

Figure 4

Expression of the PA1113 gene in the mutant PW3010 transformed with the plasmid pBBR1-Mcs5 containing the wild-type gene PA1113. RT-PCR was performed on total RNAs as described in the Materials and Methods section.

Figure 5

Effects of verapamil on sensitivity to carbenicillin of mutant PW3010 and PW3010 transformed either with pBBR5-PA1113 or the empty vector. PAO1 PW3010 mutant (♦), PW3010 transformed with plasmid pBBR5 containing the wild PA1113 gene (■), and PW3010 transformed with the empty vector pBBR5 (▲) were inoculated in 96-well plates with increasing concentrations of verapamil and carbenicillin. Growths and IC₅₀ values were measured as described in the legend of Figure 1. The IC₅₀ values were plotted against concentration in verapamil. Results presented were the mean values of at least four independent experiments.

Figure 6

Kinetics of accumulation of carbenicillin in the periplasm of P. aeruginosa PAO1 and its mutant in the PA1113 gene (PW3010) in the presence (LB + Vp) or absence (LB and PW3010) of verapamil. P. aeruginosa PAO1 suspended in LB medium or LB + 1 × 10⁻⁷ M verapamil at a concentration of 1 × 10¹⁰ bacteria per mL was incubated for 20 min at 37 °C before starting the kinetics. The kinetics and recovery of periplasmic contents were carried out as described in the Materials and Methods section. The amount of carbenicillin in periplasmic extracts was determined from a standard range established with the high sensitive bacterium Bacillus subtilis 168 and plotted against time. The results presented in the figure are the mean values of at least three independent experiments.

Figure 7

Kinetics of accumulation of carbenicillin in the cytoplasm of P. aeruginosa PAO1 in the presence (LB + Vp) or absence (LB) of verapamil and in the mutant in PA1113 gene (PW3010). P. aeruginosa PAO1 and its mutant PW3010 suspended in LB medium or LB + 1 × 10⁻⁷ M verapamil at a concentration of 1 × 10¹⁰ bacteria per mL were incubated for 20 min at 37 °C before starting the kinetics. After recovery of the periplasmic contents, PAO1 protoplasts were lysed by sonication and the cytoplasmic contents were recovered after centrifugation. The amount of carbenicillin in cytoplasmic extracts was determined from a standard range established with the high sensitive bacterium Bacillus subtilis 168 and plotted against time. The results presented in the figure are the mean values of at least three independent experiments.

Figure 8

Modeled interactions of the PA1113 ABC transporter with carbenicillin (light blue), verapamil (yellow), dibenzothiophene (green), and ATP (red). (A–D) show the general views of the binding of the PA1113 gene product to carbenicillin, verapamil, dibenzothiophene, and ATP, respectively. Detailed views of the binding site showing the principal amino acids potentially involved in the binding of carbenicillin, verapamil, dibenzothiophene, and ATP; ARG421, PHE423, ALA424 and LEU425 are implicated in binding with the different substrates.

Figure 9

Proposed mechanism for carbenicillin influx.

Figure 10

Range of sensitivity of Bacillus subtilis 168 to carbenicillin. About 1 × 10⁶ bacteria were spread on the LB-agar plate. After 1 h of incubation at 37 °C, holes were drilled in the agar with a sterile Pasteur pipette and 50 μL of a solution containing increasing quantities of carbenicillin were added to each well. After 24 h of incubation at 37 °C, the size of each halo of growth inhibition was measured and plotted as a function of the quantity of carbenicillin in the wells. Such a standard range was performed for each carbenicillin input measurement experiment in PAO1 or its mutant PW3010.